The ganglion cell degeneration observed in glaucoma (in humans and mice) is not spread diffusely over the whole retina but occurs in sectors that point towards the optic nerve head. The obvious localization for a first insult to the axons that would cause this geometry is the portion of the optic nerve head that lies directly behind the sclera, because there the axons are organized into bundles with a topographic relationship to the retina. It is possible that individual axon bundles are separately damaged or spared, leading to sectors of degenerated or normal ganglion cells in the same retina. In humans, the optic nerve head contains a rigid, collagenous lamina cribrosa, lined by astrocytes. In contrast, the murine optic nerve contains only a meshwork of astrocyes (the "glial lamina"), which is disorganized in glaucomatous nerves. Because astrocytes are not rigid, this suggests that the damage to axon bundles may not be mechanical. Instead it raises the possibility that astrocytes of the glial lamina may be directly involved in the pathophysiology of glaucoma. We propose four studies on the single-cell level of the cell biology of glial lamina astrocytes using a mouse that expresses GFP in sporadic, individual astrocytes. (1) We will describe the normal astrocytic architecture of the glial lamina. (2) We will introduce global optic nerve lesions (optic nerve crush) or focal lesions of ganglion cell axons on the retinal surface (by laser) to study the behavior of individual astrocytes under pathological conditions. (3) We will observe the consequences of these axonal lesions in transgenic mice whose astrocytes are compromised by knock-out of intermediate filaments or connexin43. (4) We will observe the behavior of GFP-labeled astrocyes in a new strain of GFP-DBA/2J glaucomatous mice as a realistic, slow model of glaucoma. PUBLIC HEALTH RELEVANCE: Glaucoma leads to a progressive and irreversible loss of retinal ganglion cells, whose axons form the optic nerve, and thereby severs the connection of an otherwise functional retina with the brain. Recent experimental evidence suggests that a non-neuronal cell type (astrocytes) in the optic nerve might play an active role in the disease. Our goal is to study the reaction of individual astrocytes to injury and glaucomatous degeneration in more detail than has been possible before.